High-Performance Computing: Dos and Don’ts

Abstract

Computational fluid dynamics (CFD) is the main field of computational mechanics that has historically benefited from advances in high-performance computing. Highperformance computing involves several techniques to make a simulation efficient and fast, such as distributed memory parallelism, shared memory parallelism, vectorization, memory access optimizations, etc. As an introduction, we present the anatomy of supercomputers, with special emphasis on HPC aspects relevant to CFD. Then, we develop some of the HPC concepts and numerical techniques applied to the complete CFD simulation framework: from preprocess (meshing) to postprocess (visualization) through the simulation itself (assembly and iterative solvers).

Fully coupled fluid‐electro‐mechanical model of the human heart for supercomputers

“…For the MPI tasks, we use an automatic mesh partition scheme using either open source METIS or an in‐house developed space filling curve (SFC) scheme . OpenMP threads are used in the MPI tasks' heavy‐duty loops following the strategy discussed in Houzeaux et al…”

“…Two bidirectionally coupled problems: electromechanical and fluid-structure one, with three systems of equations describing them: electrophysiology, solid mechanics, and fluid mechanics in deformable mesh scheme. 88 OpenMP threads are used in the MPI tasks' heavy-duty loops following the strategy discussed in Houzeaux et al 89 The coupled problem is solved following a multi-code approach, especially well-suited for staggered coupling. The simulation problem is split in two subproblems: tissue and blood inside the cavities, being each of the subproblems solved with a separate Alya instance.…”

Runtime mechanisms to survive new HPC architectures: A use case in human respiratory simulations

“…This phase consists of a loop over the elements of the mesh (see Figure 1): For each element, a set of local operations is performed to assemble the local matrices. More details can be found in Houzeaux et al (2018) and Garcia-Gasulla et al (2018a).…”

Computational Fluid and Particle Dynamics Simulations for Respiratory System

“…Synchronous n = f + p n = f' + p' Figure 3: Execution modes for CFPD simulations with Alya a set of local operations is performed in order to assemble the local matrices. More details can be found in [12,14]. From the parallelism point of view this phase has two important characteristics:…”